Sampled data and histograms have been utilized for showing data distributions in various applications. Some applications may also need to detect/find the peaks of such sampled data for various purposes. While some existing peak detection methods may be utilized in certain applications, they may be prone to errors and biases, especially when applied to noisy data. Therein lies a need for a robust peak finding method.